This pilot study is being performed in conjunction with the planning of a large, multicenter project to investigate the genetics of pancreatic cancer. The results of this study will help the planning of and direct analytic strategies in the larger multicenter study. Pancreatic cancer is the 5th leading cause of cancer death in the United States, but little is understood about its causes. Recent studies have shown that family history of cancer or family history of pancreatic cancer are possible risk factors. These data suggest a hypothesis that genetic factors are involved in pancreatic cancer susceptibility. The primary objective of this study is to determine if there is evidence for a major susceptibility gene in pancreatic cancer. (If major gene models can be delineated, they will be used in gene discovery strategies of the larger study). Segregation analysis using family data on pancreatic cancer probands will be conducted. Previously suggested risk factors, including diabetes, pancreatitis, and tobacco exposure, will be included in the analysis to determine how they modify cancer risk. Secondary objectives of this proposed study are: assessment of how ascertainment bias can affect the results of segregation analysis (an important design issue in genetic analysis of complex diseases); to determine if this bias can be adequately corrected; and to develop methods to create a sample representative of the general population from a hospital based tumor registry. Patients with pancreatic cancer have already been identified from the Johns Hopkins Hospital Tumor Registry and Johns Hopkins Bayview Medical Center and contacted by mail. A sample of these respondents that is representative of all Maryland pancreatic cancer cases will be constructed based upon characteristics identified from the Maryland Cancer Registry. Segregation analysis will be performed on the constructed sample population. Additionally, a separate segregation analysis will be performed on the families enrolled in the National Familial Pancreatic Tumor Registry, which has been developed from referrals (primarily self-referred from the Internet through a Website). This analysis will incorporate novel statistical methods to correct for the highly ascertained nature of the latter population. The two analyses will then be compared to assess the impact of ascertainment bias on segregation analysis and effectiveness of correction methods. This project thus anticipates the new analytic challenges of Internet-based subject accrual, as well as providing a basis for a large-multicenter study to understand the genetic etiology of pancreas cancer.